Plan for 24th
Recapture the beam after schottky debunch process to see if any beam remains
If beam remains, use RF shaker to knock out any remaining beam before the next cycle
Phase displacement run
Max tune measurements??
Plan to denoise schottky data:
Emi has measurements of FAB signal with noise from magnets, RF and correctors (on/off)
Emi has connected a 3rd bpm
Generated AWG scripts for adiabatic debunch and adiabatic recapture
Generated AWG scripts for abrupt debunch and adiabatic recapture
Shinji's AWG script gui does not work for abrupt debunch. (later edit: the script can be used for abrupt changes by changing voltage at the start of an interval)
Scripts were generated by splicing .equ file provided by Shinji for abrupt debunch and coast with the second half of our .equ file for adiabatic debunch/recapture
09:30:Machine setup
11:12 We will run the nominal KURNS AWG script first to make sure the machine is working as expected
11:57:machine working starting rebunching test
12:10 - RF offset timing is set to 2.110 ms
Note: While making the rebunching RF scripts I noticed that the adiabatic AWG script RF voltage only changes once phis has already reached zero in Shinji's Schottky AWG scripts. I was under the impression that the RF voltage reduces to 1kV at the same time phis goes to zero.
For the re-capture AWG scripts I have mirrored Shinji's schottky AWG scripts, see screenshot below.
The screenshot below show an example recpature AWG script with 5ms coast
12:11 - Adiabatic debunch, 5ms coast, adiabatic rebunch
Signal increases after rebunch -- recapture successful
Also took beam off data
12:28 4kV abrupt debuncg, 5ms, coast adibatic rebunch
Signal observed on FAB
Also took beam off data
12:36: 2kV abrupt debunch 5ms coast, adiabatic rebunch
Signal observed on FAB
Also took beam off data
12:41 Adiabatic debunch, 10ms coast, adiabatic rebunch
Signal seen on FAB, looks like more signal than the 5ms case
12:47 4kV, 10ms coast, adiabatic rebunch
Signal seen on FAB, less signal vs the adb case
Continue this sequnce above up to 30 ms coast, see the experiment log on the 20230124 data folder on teams for details on each set of experimental data
For a 20Hz rep rate an AWG script of around 48 ms is too long. For the 20hz 25ms and30ms coast scripts were too long. We reduced the rep rate to 10 hz and ran the 25 and 30 ms coast. With a 10 hz rep are we can use AWG scripts that are up to 100ms long
We observed signal on the FAB after rebunching particularly in the adiabtic case. So there was beam left over from the previous cycle in last week's data.
Method to remove left over beam: We have made AWG scripts for the schottky measurement which recapture and accelerate the beam to ~81 Mev and into the scraper to remove the previous beam.
30ms of coast between debunch and recapture+acceleration
RF shaker would take c. 1hr to install
15:05 begin measurement series schottky with beam killing programme (see screenshot below)
current scraper position 505mm /0055663
Loading awg scripts
RF offset changed to 2.136 ms for 10 Hz rep rate
Start taking data for schottky measurments with an RF program to kill any remaining beam
set scraper position so that beam loss is clear after recapture at t=60ms
set at original position of 505mm /0055663 - see beam loss on the FAB at this point
Recording beam loss data (i.e. fab data showing loss of beam on scraper)
In the schottky scripts with beam removal, the coasting section ie the section where we take schottky data is 30 ms, we think that should be enough to get enough resolution for the FFT
Did not see any FAB signal in the reacceleration section for the 4kv abrupt case with the 30ms of coast
15:55 Oscilloscope rebooted
Did not see any FAB signal in the reacceleration section for the 3kV case with 30ms of coast.
This morning we did see some FAB signal after 30 ms of coast and recapture in the 4kV case. The RF program from this morning recaptures into a stationary bucket. The program we are using to remove the previous beam for the schottky measurements recaptures the beam and accelerates it into the scraper at ~80MeV. When we recapture and then ramp phi_s to accelerate, the bucket area decreases. This could be the reason we do not see any signal on the FAB for the 4kV case after 30ms of coast with this afternoon's measurements.
Recaptured particles that are lost during the phi_s ramp in the reacceleration phase will continue to coast round the ring (hopeully) at a different frequency to the next beam, ie the revolution frequency peak in the schottky spectra will be uneffected (hopefully)
Finished taking schottky data with RF program to remove remaining beam
See below, adiabatic recapture data for 10 ms of coast and 30 ms of coast, both plots are the adiabatic debunch case
BEFORE DEBUNCH
This data is for the adiabtic debunch and then recapture.
Zooming into the 30 ms coast case (right hand side with_beam/no_beam plot above)
AFTER RECAPTURE
Below is the difference of the no_beam/with_beam signals in the above plots:
AFTER RECAPTURE
BEFORE DEBUNCHING
This shows that the beams survive for 30 ms, but cannot say how much more. I think it is worth measuring Schottky scan with different repetition rate up 1 Hz or as low as possible. Or we should see whehter signal from 1 Hz (or low repetition rate) operation is similar to 20 Hz operation or that with recapture. SM